Phygital marketing: repair the omnichannel journey

Phygital connects decisions, not gadgets

Phygital marketing is the coordinated use of physical and digital touchpoints within one experience. It may involve preparing a visit online, checking product information in store, resuming a quote at a branch, receiving post-purchase follow-up or transferring a case to after-sales service.

Multichannel adds entry points. Omnichannel seeks to maintain continuity between them. That continuity does not mean all data is centralized or every interaction must be recognized. It means the customer does not have to repair the system’s essential contradictions themselves.

Seven Gold recommendation. First, stop asking “which technology should we add?” Ask instead: “at which point does someone lose information, a promise, time or the ability to act?”

1. Follow a complete journey: from research to after-sales service

Consider a hypothetical specialist-retail scenario. A customer discovers a product on mobile, checks availability, visits the store, asks for advice, buys, then contacts after-sales service. Five breaks may appear:

1. Discovery: the online product page specifies neither the variant nor locally available timing.
2. Preparation: the reservation is not visible to the store team.
3. Advice: the salesperson does not know about the comparison already completed and starts the conversation again.
4. Purchase: the promise stated online differs from the condition applied at checkout.
5. After-sales service: the receipt and case are separate; the customer must explain the history again.

None of these breaks proves that an app, kiosk or augmented reality is needed. The first response may be more reliable inventory, a clear page, a shared field, a reservation identifier or a handoff protocol.

2. Map the break before the solution

Observe actual journeys, interview customers and teams, then connect every stage to an action, expectation, data item, system and owner. Note when information is requested twice, copied manually, contradicted or unavailable.

Stage	Customer decision	Observable break	Cost to document
Discovery	Does the product meet the need?	Local information is missing or different.	Exits, repeated calls, unqualified traffic.
Preparation	Is the trip worthwhile?	Stock or appointment is not confirmed.	Wasted visits, verification time.
Advice	Which option should I choose?	Online context is lost.	Re-entry, longer discussion, inconsistent recommendation.
Purchase	Can I complete now?	Price, benefit or identity is not carried over.	Abandonment, manual correction, dispute.
After-sales service	Will my problem be resolved?	History is scattered.	Delay, multiple contacts, handling error.

Choose a break that is frequent, important and controllable. A rare, spectacular irritation may deserve less priority than quiet re-entry suffered every day.

3. Build a business case before integration

Deliberately hypothetical teaching scenario. A team re-enters 500 records per month, taking eight minutes per record. That represents 4,000 minutes, or about 66.7 hours. At a fully loaded cost of €35 per hour, the theoretical time cost is approximately €2,333 per month, before errors and supervision.

This calculation is neither a client result nor a promised saving. An integration will not necessarily remove all the time; it may move the work to checking, support or exception correction. A cautious annual gain should be compared with the costs of design, licenses, development, security, migration, training, maintenance and downtime.

Hypothesis to test: automatically transferring previously validated fields reduces re-entry without increasing errors or blocking exceptional cases. Fact to measure: time per record, errors, reworked records, satisfaction and incidents before and after. Decision: deploy, correct, limit or stop.

4. Define the minimum data that provides continuity

A “360-degree customer view” is often framed as an objective even though it specifies no decision. Resuming a reservation may require only an identifier, store, product, status and expiry. The salesperson does not necessarily need the customer’s entire browsing history.

For every data item, document:

• the decision or service it enables;
• the source and system of record;
• the legal basis and information to be validated;
• authorized people and systems;
• useful retention, correction and deletion;
• behavior when the data is absent or wrong.

Regulatory fact. The GDPR notably establishes principles of purpose limitation, minimization, accuracy and limited retention for personal data. It does not prescribe your commercial architecture; it requires you to justify and govern its processing.

5. Do not confuse continuity with permanent tracking

Recognizing a record requested by the customer is not the same as tracking all their devices. If the system writes or reads information on a device, analyze the trackers involved and their purposes.

The CNIL notes that some cookies and trackers require prior consent, while strictly necessary uses may be exempt under their conditions. Refusing and withdrawing consent should be as easy as granting it.

Design the journey so it does not depend on accepted profiling: visible reservation reference, accessible receipt, option to call, trained staff. Refusing a marketing tracker should not make a promised physical service impossible when the service can operate another way.

Recommendation: have purposes, disclosures, rights, security and providers reviewed before deployment. A platform that offers a CMP or identifier does not automatically transfer responsibility.

6. A QR code is a door, not a strategy

A QR code can connect physical material to digital information: instructions, composition, availability, reservation or assistance. Its value depends on the destination, link stability, context of use and a visible alternative.

The GS1 Digital Link standard defines a way to connect GS1 identifiers to online information; URI syntax version 1.6.0 is listed as published in April 2025. The standard may be relevant for identified products. It is neither mandatory for every QR code nor a guarantee of engagement.

Before printing, test:

• the destination with imperfect connectivity and on several devices;
• URL persistence and the redirect plan;
• the promise displayed next to the code;
• actual size, contrast, lighting and distance;
• a short URL, phone number or human help as a fallback;
• useful measurement without triggering unjustified tracking.

7. Continuity includes accessibility and failure cases

A journey that requires scanning, installation, account creation or geolocation permission excludes some customers and becomes fragile as soon as the phone, network or service fails.

The W3C presents WCAG 2 as the standard for making web content more accessible, including on mobile. It concerns digital content; on its own, it does not cover the physical accessibility of a store or kiosk. The project should therefore combine web testing, on-site ergonomics and operational alternatives.

Account for the keyboard, screen reader, zoom, contrast, error comprehension and touch target, then observe use in context. A human alternative should not be hidden behind several screens.

8. Set the threshold that justifies additional technology

A new tool becomes defensible when five conditions are met:

1. the break is documented by observations and reliable data;
2. its frequency and potential cost exceed the full cost of change;
3. an owner can handle exceptions, incidents and rights requests;
4. systems of record and synchronization rules are explicit;
5. a reversible version has shown improvement without harming a guardrail.

A reversible version can be a manual protocol, structured form, shared identifier or limited synchronization. It tests the mechanism. It should not become permanent hidden debt.

Hypothesis: automation improves the handoff between two stages. Guardrails: accuracy, timing, accessibility, consent, incidents, team workload and satisfaction. If volume is low or the journey unstable, full integration can wait.

9. Manage the break, not the number of channels

The scorecard should remain tied to the chosen problem:

• Flow: eligible people, successful handoffs and drop-offs at the stage.
• Quality: synchronization errors, duplicates, corrections and unresolved cases.
• Time: customer wait, team handling and manual rework.
• Economics: margin on affected sales, service cost and the solution’s full cost.
• Trust: complaints, consent withdrawals, incidents and contextualized comments.

High scan volume may mean the printed information is missing. Low volume may signal an invisible code or an uninteresting promise. The action matters more than the technical interaction.

Prioritize the useful integration

Before adding another tool, identify the break that costs sales or time. Seven Gold can map the journey and prioritize the useful integration. Organize an omnichannel journey workshop.